app.py

# Check the requirements.txt file , import necessary libraries
import gradio as gr
import numpy as np
import cv2
from PIL import Image
from ultralytics import YOLO
from transformers import pipeline

# Loading Models via Pipeline 
yolo = YOLO("yolov8n.pt")

age_model = pipeline(
    "image-classification",
    model="nateraw/vit-age-classifier",
    device=-1
)

gender_model = pipeline(
    "image-classification",
    model="rizvandwiki/gender-classification",
    device=-1
)

# Live Detection
def live_detect(img):
    if img is None:
        return img

    frame = np.array(img)
    frame = cv2.resize(frame, (480, 360))  # mobile optimized

    results = yolo(frame, conf=0.5, classes=[0], verbose=False)[0]

    for box in results.boxes:
        x1, y1, x2, y2 = map(int, box.xyxy[0])
        crop = frame[y1:y2, x1:x2]

        if crop.size == 0:
            continue

        face = cv2.resize(crop, (96, 96))
        pil_face = Image.fromarray(face)

        age = age_model(pil_face)[0]["label"]
        gender = gender_model(pil_face)[0]["label"]

        label = f"{gender}, {age}"

        cv2.rectangle(frame, (x1, y1), (x2, y2), (0, 255, 0), 2)
        cv2.putText(
            frame,
            label,
            (x1, y1 - 6),
            cv2.FONT_HERSHEY_SIMPLEX,
            0.5,
            (0, 255, 0),
            2,
        )

    return Image.fromarray(frame)

# GradIO User Interface (UI)
with gr.Blocks() as demo:

    gr.Markdown(
        """
        Human Lens  :: Live Detection with Gender & Age
        """
    )

    cam = gr.Image(
        sources=["webcam"],
        type="pil",
        streaming=True,
        label="Live Camera",
        elem_id="cam"
    )

    cam.stream(
        fn=live_detect,
        inputs=cam,
        outputs=cam,
        show_progress=False
    )

    gr.Markdown("<hr><center><b>Developed by Phani</b></center>")

demo.queue().launch(
    css="""
    #cam {max-width:100%; border-radius:12px;}
    """
)